STEP 01
Dendrite
Branching antennae collect incoming chemical messages from thousands of neighbouring neurons.
Understanding Intelligence Beyond Boundaries
Where Biology Meets Binary —
Where Biology Meets Binary —
Intelligence Beyond Boundaries
Eighty-six billion neurons of wetware, versus billions of tuned parameters of silicon. Explore how the human brain thinks — and how machines learned to imitate it.
Human Brain · 20 Watts
Neural Network · Megawatts
About the Human Brain
Roughly 1.4 kg of tissue that composes symphonies, proves theorems and dreams — on less power than a light bulb. Here is how it is built, how it signals, and what makes it astonishing.
01 — Structure of the Brain
Hover a region on the diagram, or tap a label button, to reveal what it does.
Select a region
Frontal Lobe
Click any label above or hover the diagram to read about that part of the brain.
02 — How Human Neurons Communicate
Watch the signal travel: Dendrite → Cell Body → Axon → Synapse → Next Neuron.
STEP 02
Cell Body
The soma sums every excitatory and inhibitory input. Cross the threshold, and it fires.
STEP 03
Axon
An action potential races down the myelinated axon at up to 120 metres per second.
STEP 04
Synapse
The electrical spike becomes chemical: neurotransmitters cross a 20-nanometre gap.
STEP 05
Next Neuron
Receptors reopen the cycle — and each repetition quietly rewires the connection's strength.
03 — Amazing Facts About the Human Brain
Hover (or focus) a card to flip it.
100B
Neurons
Around 86–100 billion neurons, wired by roughly 100 trillion synapses — more connections than stars in the Milky Way.
60%
Fat Content
The brain is the fattiest organ in the body. Lipids build myelin and cell membranes, keeping signals fast and insulated.
120M
Images / Sec
Visual cortex throughput is often cited near 120 million image-forming signals per second — vision alone uses ~30% of the cortex.
24/7
Never Stops
It never switches off. During sleep it consolidates memories, flushes waste and rehearses the day at higher intensity than some waking states.
23 W
Power Output
About 20–23 watts — roughly a dim light bulb — yet it outperforms data centres at general reasoning and adaptation.
1M GB
Storage
Estimated capacity near one petabyte — about 1 million gigabytes, or three million hours of television.
Discover Artificial Intelligence
The silicon counterpart: artificial neurons, trained weights and attention — inspired by biology, engineered by mathematics.
Artificial Neurons
A weighted sum plus a non-linear activation — a cartoon of a biological neuron, repeated billions of times.
Backpropagation
Machines learn by measuring error and nudging every weight backwards. Brains learn locally, without a global error signal.
Deep Networks
Stacked layers build hierarchy: edges to shapes to objects to meaning — echoing the visual cortex.
The Gap
AI wins on scale and recall. The brain wins on energy, generalisation and learning from a handful of examples.
Human Brain vs Artificial Intelligence
Two very different kinds of intelligence. Here is how they really compare, side by side.
Aspect
Human Brain
Artificial Intelligence
Hardware
~86 billion biological neurons, ~100 trillion synapses
Silicon chips (GPUs/TPUs) running billions of numeric parameters
Energy use
About 20 watts — less than a light bulb
Kilowatts to megawatts for training large models
Learning
Learns from a handful of examples; local, continuous, lifelong
Needs massive datasets; learns via backpropagation in training runs
Memory
Associative, reconstructive and imperfect — but meaningful
Exact storage and recall; limited context window while reasoning
Speed
Signals travel up to ~120 m/s; massively parallel but slow per step
Billions of operations per second; extremely fast arithmetic
Creativity
Genuine imagination shaped by emotion, culture and experience
Recombines learned patterns; novelty without lived experience
Emotion & consciousness
Feels, wants, suffers, is self-aware
Simulates emotional language; no inner experience
Errors
Biased, forgetful, tires — but knows uncertainty intuitively
Confidently wrong (hallucination); no true understanding of limits
Adaptability
Transfers skills across totally new situations
Strong inside its training distribution, brittle outside it
Scaling
Cannot be copied, upgraded or run in parallel
Copied instantly, scaled to millions of users at once
Where the brain wins
Common sense, empathy, moral judgement, learning from one example and staggering energy efficiency.
Where AI wins
Perfect recall, tireless speed, superhuman pattern search across data no human could ever read.
Better together
Humans set goals and meaning; machines handle scale and repetition. Augmentation beats replacement.
Sign In
Sign in with your email to attempt the quiz — your points are saved to that email on this device.
Brain & AI Quiz
Ten questions · 10 points each · answer them all to bank your score.
You need to sign in before attempting the quiz, so your points can be saved to your email.
Go to sign inSaved points by email
No scores saved yet.
The Future of Intelligence
From neurons to silicon — and beyond. A glimpse at where the human mind and artificial intelligence are headed together.
2025
Brain-Computer Interfaces
Implantable BCIs allow direct neural-to-digital communication — letting paralysed patients type with thought alone, blurring the border between mind and machine.
2027
Neuromorphic Computing
Chips that mimic spiking neurons fire only when needed — slashing energy use by 1000× versus GPUs. AI that thinks more like a brain, at brain-like power.
2030
Whole-Brain Emulation
Mapping every synapse at nanometre resolution — the connectome — opens a path to digital twins of real brains, enabling unprecedented neuroscience and AI architectures.
2035+
Augmented General Intelligence
Not AI replacing humans — but AI seamlessly extending human cognition. Memory, creativity and analysis amplified, while consciousness remains irreducibly human.
Key Takeaways
Everything explored — distilled into four ideas worth remembering.
Efficiency Is the Brain's Superpower
Running on 20 W, the brain outperforms any silicon system on general reasoning, adaptation and learning from a handful of examples — a gap engineers are still chasing.
AI Wins on Scale & Speed
Where sheer throughput, perfect recall and exhaustive search across billions of data points matter, no human can compete — and that's genuinely useful.
Collaboration > Competition
The most powerful intelligence systems of the next decade will be human-AI teams: humans set values and goals; machines handle execution and pattern recognition.
Consciousness Remains Unsolved
The hard problem — why physical processes give rise to subjective experience — is still open. Until it's answered, calling AI "intelligent" in the full human sense is premature.
Ready to Test Your Knowledge?
How Well Do You Know
How Well Do You Know
Intelligence?
You've explored the biology, the silicon, the comparisons and the future. Now put it all to the test — ten questions, 100 points, and your score saved instantly.
86BNeurons in
your brain
your brain
20WPower it
consumes
consumes
1PBEstimated
storage
storage
∞Potential
beyond limits
beyond limits